[asterisk-bugs] [Asterisk 0012781]: Queue holdtime smoothing filter is not boxcar (FIR), but exponential average (IIR).

[noreply at bugs.digium.com]

A NOTE has been added to this issue. 
====================================================================== 
http://bugs.digium.com/view.php?id=12781 
====================================================================== 
Reported By:                davidw
Assigned To:                
====================================================================== 
Project:                    Asterisk
Issue ID:                   12781
Category:                   Applications/app_queue
Reproducibility:            always
Severity:                   text
Priority:                   normal
Status:                     new
Asterisk Version:           1.4.20.1 
SVN Branch (only for SVN checkouts, not tarball releases): N/A 
SVN Revision (number only!):  
Disclaimer on File?:        N/A 
Request Review:              
====================================================================== 
Date Submitted:             06-03-2008 13:01 CDT
Last Modified:              06-13-2008 13:35 CDT
====================================================================== 
Summary:                    Queue holdtime smoothing filter is not boxcar (FIR),
but exponential average (IIR).
Description: 
The comment for recalc_holdtime says that it implements a recursive boxcar
algorithm.  Even with the "recursive" qualifier, boxcar averaging refers to
simple rolling average filters, which are finite impulse response ones. 
The filter that recalc_holdtime actually implements is an exponential
average, which is an infinite impulse response filter.


====================================================================== 

---------------------------------------------------------------------- 
 putnopvut - 06-13-08 13:35  
---------------------------------------------------------------------- 
I understand that the comment and code do not match, but I guess there
needs to be a distinction made regarding this report. Does the comment need
to be changed to conform to what's in the code or does the code need to be
changed to conform to what the comment says? Unfortunately, I don't really
grasp the gain from changing the code to be a FIR function so my initial
reaction upon seeing such a report is to change the comment. I didn't write
the holdtime calculation function, but looking at it, I have to believe it
was written the way it was to be lightweight. 

Issue History 
Date Modified   Username       Field                    Change               
====================================================================== 
06-13-08 13:35  putnopvut      Note Added: 0088693                          
======================================================================